REGRESSION (r70335): Incorrect article layout in Safari Reader
Created attachment 72054 [details] Disable the cross-origin check in getMatchedCSSRules in versions of Safari lined against versions of WebKit that didn’t have the check
Comment on attachment 72054 [details] Disable the cross-origin check in getMatchedCSSRules in versions of Safari lined against versions of WebKit that didn’t have the check View in context: https://bugs.webkit.org/attachment.cgi?id=72054&action=review > WebCore/ChangeLog:9 > + (WebCore::DOMWindow::getMatchedCSSRules): Return rules from all origins if the corss-origin Typo: corss > WebCore/page/DOMWindow.cpp:1257 > Document* doc = m_frame->document(); This local variable doesn’t seem to keep its weight. > WebCore/page/Settings.h:336 > + bool disableGetMatchedCSSRulesCrossOriginCheck() const { return m_disableGetMatchedCSSRulesCrossOriginCheck; } For functions that return booleans, I prefer that we not use verb phrases such as “disable”. > WebCore/page/Settings.h:425 > + bool m_disableGetMatchedCSSRulesCrossOriginCheck : 1; Same thing for data members.
Comment on attachment 72054 [details] Disable the cross-origin check in getMatchedCSSRules in versions of Safari lined against versions of WebKit that didn’t have the check View in context: https://bugs.webkit.org/attachment.cgi?id=72054&action=review >> WebCore/page/DOMWindow.cpp:1257 >> Document* doc = m_frame->document(); > > This local variable doesn’t seem to keep its weight. Sorry, I meant to say “doesn’t improve readability or clarity and is only used once”.
Fixed in <http://trac.webkit.org/changeset/70714>.
Marking as resolved per the above.
<rdar://problem/8591336>